LLM增强的网络社交智能体模拟
原标题：LLM-Augmented Agent-Based Simulation of Cyber Social Agents
AI 摘要
该论文提出「网络社交智能体」（Cyber Social Agents, CSA）概念，用原型分类法按操作策略与修辞风格区分社交媒体自动化账号，而非简单二分「机器人/人类」。作者将这一分类法整合进由 LLM 驱动的社交模拟，用基于实证数据构建的异构 CSA 人格在网络环境中通信与协作，研究智能体多样性如何影响信息扩散与社会影响力。工作旨在弥合规则型智能体模型与 LLM 生成式模拟之间的差距。
正文节选
LLM-Augmented Agent-Based Simulation of Cyber Social Agents Abstract Social media bots are often portrayed as malicious automations to be detected and removed, but automation on social media platforms encompasses a wide spectrum of roles, behaviors, and rhetorical strategies that can shape online discourse in both constructive and destructive ways. This work conceptualizes these automated actors as Cyber Social Agents through a taxonomy of archetypes characterized by their operational tactics a